A new digital elevation model of Antarctica derived from CryoSat-2 altimetry - The Cryosphere
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
The Cryosphere, 12, 1551–1562, 2018
https://doi.org/10.5194/tc-12-1551-2018
© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.
A new digital elevation model of Antarctica derived from
CryoSat-2 altimetry
Thomas Slater1 , Andrew Shepherd1 , Malcolm McMillan1 , Alan Muir2 , Lin Gilbert2 , Anna E. Hogg1 ,
Hannes Konrad1 , and Tommaso Parrinello3
1 Centre for Polar Observation and Modelling, School of Earth and Environment, University of Leeds,
Leeds, LS2 9JT, UK
2 Centre for Polar Observation and Modelling, University College London, London, WC1E 6BT, UK
3 ESA ESRIN, Via Galileo Galilei, 00044 Frascati RM, Italy
Correspondence: Thomas Slater (py10ts@leeds.ac.uk)
Received: 3 October 2017 – Discussion started: 25 October 2017
Revised: 6 March 2018 – Accepted: 22 March 2018 – Published: 2 May 2018
Abstract. We present a new digital elevation model (DEM) et al., 2015). Accurate knowledge of surface elevation can
of the Antarctic ice sheet and ice shelves based on 2.5 × 108 be used for both the delineation of drainage basins and es-
observations recorded by the CryoSat-2 satellite radar al- timation of grounding line ice thickness, necessary for es-
timeter between July 2010 and July 2016. The DEM is timates of Antarctic mass balance calculated via the mass
formed from spatio-temporal fits to elevation measurements budget method (Rignot et al., 2011b; Shepherd et al., 2012;
accumulated within 1, 2, and 5 km grid cells, and is posted Sutterly et al., 2014). Furthermore, detailed and up-to-date
at the modal resolution of 1 km. Altogether, 94 % of the DEMs are required to distinguish between phase differences
grounded ice sheet and 98 % of the floating ice shelves are caused by topography and ice motion when estimating ice
observed, and the remaining grid cells north of 88◦ S are in- velocity using interferometric synthetic aperture radar (Rig-
terpolated using ordinary kriging. The median and root mean not et al., 2011a; Mouginot et al., 2012).
square difference between the DEM and 2.3 × 107 airborne Previously published DEMs of Antarctica have been de-
laser altimeter measurements acquired during NASA Oper- rived from satellite radar altimetry (Helm et al., 2014; Fei
ation IceBridge campaigns are −0.30 and 13.50 m, respec- et al., 2017), laser altimetry (DiMarzio et al., 2007), a combi-
tively. The DEM uncertainty rises in regions of high slope, nation of both radar and laser altimetry (Bamber et al., 2009;
especially where elevation measurements were acquired in Griggs and Bamber, 2009), and the integration of several
low-resolution mode; taking this into account, we estimate sources of remote sensing and cartographic data (Liu et al.,
the average accuracy to be 9.5 m – a value that is comparable 2001; Fretwell et al., 2013). In addition, high-resolution re-
to or better than that of other models derived from satellite gional DEMs of the marginal areas of the ice sheet have been
radar and laser altimetry. generated from stereoscopic (Korona et al., 2009) and ra-
diometer surveys (Cook et al., 2012). Although these pho-
togrammetric models perform well over regions of bare rock
and steep slope found in the margins, their accuracy is con-
1 Introduction siderably reduced in ice-covered areas.
CryoSat-2, launched in 2010, is specifically designed to
Digital elevation models (DEMs) of Antarctica are important overcome the challenges of performing pulse-limited altime-
data sets required for the planning of fieldwork, numerical try over Earth’s polar regions. With a high inclination, drift-
ice sheet modelling, and the tracking of ice motion. Mea- ing orbit, and novel instrumentation which exploits inter-
surements of ice sheet topography are needed as a bound- ferometry to obtain high-spatial-resolution measurements in
ary condition for numerical projections of ice dynamics and areas of steep terrain, CryoSat-2 provides a high-density
potential sea level contributions (Cornford et al., 2015; Ritz
Published by Copernicus Publications on behalf of the European Geosciences Union.
1552 T. Slater et al.: A new digital elevation model of Antarctica derived from CryoSat-2 altimetry
network of elevation measurements up to latitudes of 88◦ evation measurements recorded in LRM are slope-corrected
(Wingham et al., 2006). Here, we utilise a 6-year time se- by relocating the echoing point away from nadir in accor-
ries of elevation measurements acquired by CryoSat-2 be- dance with the surface slope, determined using an external
tween July 2010 and July 2016 to derive a comprehensive DEM (Radarsat Antarctic Mapping Project version 2 DEM,
and contemporary DEM of Antarctica at a spatial resolu- posted at 200 m) (Liu et al., 2001; ESA, 2012). SARIn ac-
tion of 1 km, with high data coverage in both the ice sheet quisitions are slope-corrected using the interferometric phase
interior and its complex marginal areas. We then evaluate difference calculated at the location of the elevation measure-
the accuracy of the generated DEM against a set of contem- ment. For the ice shelves, additional inverse barometric and
poraneous airborne laser altimeter measurements, obtained ocean tide corrections are also applied.
during NASA Operation IceBridge campaigns, in several lo- Within the LRM mode mask area we select Level 2 ele-
cations covering Antarctica’s ice sheet and ice shelves. The vation estimates retrieved using the Offset Centre of Gravity
DEM we describe here features several improvements over (OCOG) retracking algorithm, which defines a rectangular
the preliminary ESA CryoSat-2 Antarctic DEM distributed box around the centre of gravity of an altimeter waveform
in March 2017 and should be used in its place. These im- based upon its power distribution (Wingham et al., 1986).
provements include an increase in resolution from 2 to 1 km, The OCOG retracking point is taken to be the point on the
an increase in data coverage on the grounded ice sheet from leading edge of the waveform which first exceeds 30 % of
91 to 94 %, and the use of a more robust ordinary kriging the rectangle’s amplitude (Davis, 1997). We use the OCOG
interpolation scheme to provide a continuous elevation data retracker as it offers robust retracking over a wide range of
set. surfaces and is adaptable to a variety of pulse shapes (Wing-
ham et al., 1986; Davis, 1997; Armitage et al., 2014). For the
SARIn area, where CryoSat-2 operates as a SAR altimeter
2 Data and methods and waveform characteristics differ from those acquired in
LRM, elevations are retrieved using the ESA Level 2 SARIn
2.1 CryoSat-2 elevation measurements retracker, which determines the retracking correction from
fitting the measured waveform to a modelled SAR waveform
We use 6 years of CryoSat-2 Baseline-C Level 2 measure- (Wingham et al., 2006; ESA, 2012). Over the ice sheet and
ments of surface elevation recorded by the SIRAL (SAR In- ice shelves, we use approximately 2.5 × 108 CryoSat-2 ele-
terferometer Radar Altimeter) instrument, mounted on the vation measurements to derive the new DEM.
CryoSat-2 satellite, between July 2010 and July 2016. Over
Antarctica, SIRAL samples the surface in two operating 2.2 DEM generation
modes: low-resolution mode (LRM) and synthetic aperture
radar interferometric mode (SARIn). In LRM, CryoSat-2 op- To compute elevation, we separate the input CryoSat-2 ele-
erates as a conventional pulse-limited altimeter (Wingham vation measurements into approximately 1.4 × 107 regularly
and Wallis, 2010), illuminating an area of approximately spaced 1 km2 geographical regions. We then use a model
2.2 km2 , with an across track width of roughly 1.5 km. LRM fit method to separate the various contributions to the mea-
is used in the interior of the ice sheet, where low slopes and sured elevation fluctuations within each region (Flament
homogenous topography on the footprint scale are generally and Remy, 2012; McMillan et al., 2014). This method best
well suited for pulse-limited altimetry. suits CryoSat-2’s 369-day orbit cycle, which samples along
In SARIn, SIRAL uses two receiver antennae to perform a dense network of ground tracks with few coincident re-
interferometry, allowing the location of the point of closest peats. We model the elevation Z (Eq. 1) as a quadratic func-
approach (POCA) to be precisely determined in the across- tion of local surface terrain (x, y), a time invariant term h
track plane (Wingham et al., 2004). Bursts of 64 pulses are accounting for anisotropy in radar penetration depth depend-
emitted at a high pulse repetition frequency, and Doppler pro- ing on satellite direction (Armitage et al., 2014), and a linear
cessing is then used to reduce the along-track footprint to ap- rate of elevation change with time t. The satellite heading
proximately 300 m (Wingham et al., 2006). This increased term, h, is a binary term set to 0 or 1 for an ascending or
sampling density and ability to calculate the along- and descending pass, respectively.
across-track location of the POCA make SARIn well suited Z (x, y, t, h) = z + a0 x + a1 y + a2 x 2 + a3 y 2
for measuring the steep and complex topography found in the
+ a4 xy + a5 (h) + a6 (t − tJuly 2013 ) (1)
ice sheet margins.
The CryoSat-2 Level 2 elevation product has a series of We retrieve the model coefficients in each grid cell using
geophysical corrections applied to correct the selected mea- an iterative least-squares fit to the observations to minimise
surements for the following: off-nadir ranging due to slope, the impact of outliers, and we discard unrealistic estimates
dry atmospheric propagation, wet atmospheric propagation, resulting from poorly constrained model fits (see Sect. S1
ionosphere propagation, solid-earth tide, and ocean loading of Supplement). At a spatial resolution of 1 km this approach
tide (ESA, 2012). As part of the Level 2 processing chain, el- provides, on average, in excess of 30 elevation measurements
The Cryosphere, 12, 1551–1562, 2018 www.the-cryosphere.net/12/1551/2018/
T. Slater et al.: A new digital elevation model of Antarctica derived from CryoSat-2 altimetry 1553
Figure 1. Area coverage of elevation values provided by the model
fit solution (Eq. 1) of CryoSat-2 measurements for the Antarctic ice
sheet, with a grid cell sizing of 1, 2, and 5 km2 . Solid black lines
and numbers (inset) show the boundaries and ID numbers of the
27 drainage basins used (Zwally et al., 2012). East Antarctica and
the Antarctic Peninsula are defined as numbers 2 to 17 and 24 to 27,
respectively, and the remaining numbers define West Antarctica.
per grid cell to constrain each solution. By using the model
fit method, we are able to generate elevation estimates from
6 years of continuous CryoSat-2 data, which are not unduly
affected by fluctuations in surface elevation that may occur Figure 2. The grid cell resolution of the model fit method used to
derive the surface elevation in each 1 km grid cell. Elevation val-
during the acquisition period (McMillan et al., 2014). In ad-
ues obtained from the 2 and 5 km model fits are oversampled to the
dition, it also allows for the retrieval of ice sheet elevation modal DEM resolution of 1 km. A black grid cell denotes a cell that
and rate of elevation change from the same data in a self- contains an interpolated value. For the grounded ice sheet, approx-
consistent manner. imately 60, 30, and 5 % of elevation values are derived from 1, 2,
We form the DEM from the mean elevation term, z, in and 5 km model fits, respectively. For the ice shelves, 75 % of eleva-
Eq. (1) within each 1 × 1 km grid cell, which corresponds to tions are calculated with 1 km model fits, and 23 % from 2 km model
the elevation at the midpoint of the observation period. At fits. The remaining 5 % of ice sheet and 2 % of ice shelf values
a resolution of 1 km, the model fit provides an elevation es- are interpolated using ordinary kriging. At the mode mask bound-
timate in 60 and 75 % of grid cells within the total area of ary, where CryoSat-2 switches between LRM and SARIn operating
the ice sheet and ice shelves, respectively. To fill data gaps modes, grid cells are predominantly derived from 2 km model fits,
in the 1 km grid, we generate additional DEMs of Antarc- as there are a reduced number of elevation measurements available
to constrain model fits at a resolution of 1 km.
tica from model fits at spatial resolutions of 2 and 5 km. At
these coarser resolutions more data are available to constrain
model fits within a given geographical region, particularly
at lower latitudes where the spacing between ground tracks
is larger. As a result, for the ice sheet the data coverage for using ordinary kriging (Isaaks and Srivastava, 1989; Kitani-
DEMs generated at resolutions of 2 and 5 km is increased to tis, 1997), an interpolation technique used in the generation
91 and 94 %, respectively (Fig. 1). For the ice shelves we use of previously published DEMs (Bamber et al., 2009; Helm
an additional DEM generated from model fits at a resolu- et al., 2014). We interpolate using a search radius of 10, 25, or
tion of 2 km, for which data coverage is increased to 98 %. 50 km, depending on which first satisfies a minimum thresh-
Data gaps in the 1 km grid are filled by the re-sampled 2 old of 100 data points to be used in the interpolation. Over
and 5 km DEMs (where neither 1 km or 2 km model fit es- the grounded ice sheet, 44, 52 and 4 % of interpolated eleva-
timates are available) for the ice sheet, and the 2 km DEM tion values used a search radius of 10, 25, and 50 km, respec-
for the ice shelves (Fig. 2). This approach provides a DEM tively. The majority of data points requiring a search radius
at the modal spatial resolution of 1 km, where approximately of 50 km are located along the margins of Graham Land and
94 and 98 % of grid cells contain an elevation estimate de- Palmer Land in the Antarctic Peninsula, where data coverage
rived from CryoSat-2 measurements for the ice sheet and ice is poor. After interpolation, the DEM provides a continuous
shelves, respectively. elevation data set for the ice shelves and ice sheet for lati-
In order to provide a continuous data set, we estimate ele- tudes north of 88◦ S. We have chosen not to interpolate the
vation values in grid cells north of 88◦ S that contain no data pole hole due to interpolation distances exceeding the maxi-
www.the-cryosphere.net/12/1551/2018/ The Cryosphere, 12, 1551–1562, 2018
1554 T. Slater et al.: A new digital elevation model of Antarctica derived from CryoSat-2 altimetry
mum kriging search radius of 50 km and a desire to keep the
DEM a product of CryoSat-2 data only.
2.3 Airborne elevation measurements
To evaluate the accuracy of the DEM, we compare our ele-
vation estimates to measurements acquired by airborne laser
altimeters during NASA’s Operation IceBridge survey. The
IceBridge mission, running since 2009, is the largest air-
borne polar survey ever undertaken (Koenig et al., 2010).
The primary goal of IceBridge is to maintain a continuous
time series of laser altimetry over the Arctic and the Antarc-
tic, bridging the gap between the Ice, Cloud, and Land El-
evation Satellite (ICESat), which stopped collecting data in
2009, and ICESat-2, planned for launch in 2018.
We compare the DEM to elevation measurements obtained
by two airborne laser altimeter instruments (Fig. 3):
– The Airborne Topographic Mapper (ATM), over the fol-
lowing regions of the continental ice sheet: Antarctic
Peninsula; Bellingshausen, Amundsen, and Getz sectors
of West Antarctica; and the Transantarctic Mountains,
Oates Land, and the plateau region of East Antarctica.
The following ice shelves were also surveyed: Larsen
Figure 3. IceBridge airborne data set used to evaluate the DEM,
C, Pine Island, Thwaites, Wilkins, Abbot, Getz, George
acquired between December 2008 and December 2014. The mode
VI, Ross, and Filchner-Ronne. Measurements were ac- mask boundary (solid black line) between CryoSat-2 LRM and
quired between March 2009 and December 2014 (Kra- SARIn modes is also shown. (Inset) Locations of the individual
bill, 2016). ATM and RLA airborne data sets. Labelled are the following lo-
cations of interest: AbIS: Abbot Ice Shelf; AS: Amundsen Sea; BS:
– The Riegl Laser Altimeter (RLA), over the Antarc- Bellingshausen Sea; BG: Byrd Glacier; DC: Dome C; DML: Dron-
tic Peninsula; Marie Byrd Land of West Antarctica; ning Maud Land; FIS: Foundation Ice Stream; FR: Filchner-Ronne
Dronning Maud Land, Totten Glacier, and Wilkes Land Ice Shelf; G: Getz; GL: Graham Land; GVI: George VI Ice Shelf;
of East Antarctica; and the Ross Ice Shelf. Measure- GVL: George V Land; LC: Larsen-C Ice Shelf; LD: Law Dome;
ments were acquired between December 2008 and Jan- LV: Lake Vostok; MBL: Marie Byrd Land; OL: Oates Land; PIG:
uary 2013 (Blankenship et al., 2013). Pine Island Glacier; PL: Palmer Land; PM: Pensacola Mountains;
RG: Recovery Glacier; RIS: Ross Ice Shelf; TG: Totten Glacier;
The ATM is an airborne scanning laser altimeter capable ThG: Thwaites Glacier; TM: Transantarctic Mountains; VL: Victo-
of measuring surface elevation with an accuracy of 10 cm ria Land; WIS: Wilkins Ice Shelf; WL: Wilkes Land.
or better (Krabill et al., 2004). Flown at a typical altitude
of 500 m a.g.l., the ATM illuminates a swath width of ap-
proximately 140 m, with a footprint size of 1–3 m and along- changes in elevation resulting from the laser altimeter rang-
track separation of 2 m (Levinsen et al., 2013). Data acquired ing from cloud cover (Young et al., 2008; Kwok et al., 2012).
by the RLA were collected as part of the NASA Investi-
gating the Cryospheric Evolution of the Central Antarctic 2.4 DEM evaluation
Plate (ICECAP) programme from December 2009 to 2013,
mounted to a survey aircraft flown at a typical height of When comparing the DEM and airborne laser altimeter data
800 m. Elevation measurements are provided at a spatial res- sets, we separate the evaluation results according to whether
olution of 25 m along track and 1 m across track with an error the IceBridge elevation measurement resides in a grid cell de-
of approximately 12 cm (Blankenship et al., 2013). rived from CryoSat-2 surface height measurements, hereby
In total, we selected approximately 2.3 × 107 laser al- referred to as an observed grid cell, or an interpolated eleva-
timeter elevation measurements, comprising 1.7 × 107 ATM tion value. This approach allows the accuracy of CryoSat-2
measurements and 0.6×107 RLA measurements. Combined, observations and the chosen interpolation method to be as-
these data provide an independent comparison data set, ob- sessed independently. In total, approximately 84 % of the
tained over a contemporaneous time period and in a wide airborne laser elevation measurements reside within an ob-
range of locations across Antarctica. For all airborne mea- served DEM grid cell. Of this total, 53, 41, and 6 % grid cells
surements, a filter was applied to remove any erroneous step are derived from 1, 2, and 5 km model fits, respectively.
The Cryosphere, 12, 1551–1562, 2018 www.the-cryosphere.net/12/1551/2018/
T. Slater et al.: A new digital elevation model of Antarctica derived from CryoSat-2 altimetry 1555
Figure 4. (a) A new elevation model of Antarctica derived from 6 years of CryoSat-2 radar altimetry data acquired between July 2010 and
July 2016, and (b) uncertainty map of the new CryoSat-2 Antarctic DEM, calculated from the root mean square difference of elevation
residuals in observed grid cells, and the kriging variance error in interpolated grid cells. Uncertainties due to radar penetration into a dry
snowpack are not accounted for.
In order to compare the DEM and IceBridge data sets, pographic undulations and identify the ice divides and larger
we estimate the DEM elevation at the exact location of the features such as subglacial Lake Vostok.
airborne laser altimeter measurement through bilinear inter- To evaluate the DEM’s systematic bias, we compute the
polation. Subsequently, we subtract the IceBridge elevation median elevation difference with respect to the airborne mea-
from the interpolated DEM elevation and collate the eleva- surements, as this is robust against the effect of outliers. To
tion differences into the same 1 × 1 km grid that the DEM evaluate its random error, we calculate the root mean square
is projected on. We then calculate the median difference to (rms) difference. Both of these statistical measures are more
obtain one elevation difference for each individual grid cell appropriate than the mean and standard deviation when de-
and to minimise the impact of outliers. On average, 1 km scribing the systematic bias and random error, respectively,
DEM grid cells overflown by IceBridge campaigns contain of the non-Gaussian distributions we typically find when cal-
70 individual airborne measurements. In total, elevation dif- culating elevation differences between the DEM and Ice-
ferences were compared for approximately 2.7 × 105 DEM Bridge elevation data sets.
grid cells, covering 2 % of the total ice sheet and ice shelf
area. All DEM and IceBridge elevations are referenced to 3.1 Comparison of DEM to airborne elevation
the WGS84 ellipsoid. measurements: observed grid cells
A primary objective of NASA’s IceBridge programme is to
3 Results maintain a continuous observational record of rapidly chang-
ing areas in Antarctica. As a result, elevation measurements
Our new DEM of Antarctica (Fig. 4) provides an elevation were obtained in regions such as Pine Island (PIG), Thwaites,
value derived from CryoSat-2 measurements for 94 % of the and Totten glaciers, where the observed thinning rate is of
grounded ice sheet and 98 % of the ice shelves. The remain- the order of several metres per year (McMillan et al., 2014).
ing 5 % of grid cells north of 88◦ S are interpolated using or- Therefore, we expect to see height differences between the
dinary kriging to provide a continuous gridded elevation data DEM and airborne data sets due to real changes in surface
set for the entire continent beyond the pole hole. Accounting elevation between their respective acquisition periods. When
for the length of the elevation time series within each indi- comparing DEM elevations at PIG against measurements ac-
vidual grid cell, we determine the effective time stamp of the quired by ATM flights in the years 2009, 2011, and 2014
DEM to be July 2013. Surface slopes derived from the eleva- (Fig. 6), the elevation difference is smallest in 2014, closest
tion gradient of the DEM (Fig. 5) illustrate the short-scale to- to the DEM effective date of July 2013 (Table 1).
www.the-cryosphere.net/12/1551/2018/ The Cryosphere, 12, 1551–1562, 2018
1556 T. Slater et al.: A new digital elevation model of Antarctica derived from CryoSat-2 altimetry Figure 5. (a) Surface slopes of Antarctica posted at a resolution of 1 km, derived from the digital elevation model, and (b) estimated uncertainty of surface slope, derived through propagation of the elevation uncertainties. The mode mask boundary between CryoSat-2 LRM and SARIn modes is also shown in white. Figure 6. Difference between observed DEM grid cells derived from 1 km model fits and IceBridge ATM elevation measurements for the Pine Island Glacier region in West Antarctica for ATM flight surveys undertaken in the years (a) 2009, (b) 2011, and (c) 2014. The DEM has an effective time stamp of July 2013. The boundary of the Pine Island Glacier drainage basin (solid black line) is also shown (Zwally et al., 2012). To account for the temporal difference between the two ations in backscatter, which can introduce spurious signals data sets, we adjust the interpolated DEM value for changes in time series of elevation change, by adjusting the eleva- in surface elevation which may have occurred between the tion time series according to the correlation between changes acquisition periods. We calculate this adjustment by interpo- in elevation and backscattered power (see Sect. S2) (Wing- lating the gridded elevation trends (Eq. 1) to the location of ham et al., 1998; Davis and Ferguson, 2004; Khvorostovsky, the airborne measurement, through the same bilinear interpo- 2012). lation method as used for the DEM elevation estimate. The At the continental scale, there is generally good agreement elevation change trends were corrected for temporal fluctu- between the DEM and airborne laser altimeter measurements The Cryosphere, 12, 1551–1562, 2018 www.the-cryosphere.net/12/1551/2018/
T. Slater et al.: A new digital elevation model of Antarctica derived from CryoSat-2 altimetry 1557
Table 1. Statistics of the comparison between observed DEM grid
cells derived from 1 km model fits and ATM elevation measure-
ments in the Pine Island Glacier drainage basin. The airborne data
are separated into the year of acquisition to demonstrate the effect
of ice dynamical thinning in this region on the elevation difference.
The effective date of the DEM is July 2013.
Year Number of Median rms difference
compared difference (m)
grid cells (m)
2009 3490 −4.97 11.35
2011 2819 −2.08 8.56
2014 2794 0.05 3.97
(Figs. 7 and 8), and the median and rms elevation difference
between the DEM and airborne data are −0.27 and 13.36 m,
respectively (Table 2).
At the Antarctic Peninsula, the median and rms difference
are −1.12 and 22.40 m, respectively; errors are larger in this
region due to its mountainous and highly variable terrain, and
it remains a challenge for radar altimetry. The largest eleva-
tion differences in this region are found in DEM grid cells
derived from 5 km model fits, indicating that the complex to- Figure 7. Difference between CryoSat-2 DEM elevation and air-
pography is poorly described by a quadratic model at this res- borne laser altimeter measurements in observed grid cells. The
olution. In grid cells with elevation values derived from 1 km mode mask boundary between CryoSat-2 LRM and SARIn modes
model fits, which account for 40 % of the Antarctic Penin- is also shown as a solid black line. (Inset) distribution of the eleva-
sula DEM, the median and rms difference are improved to tion differences (DEM–airborne) for the ice sheet and ice shelves.
−0.71 and 16.88 m, respectively. Geographically, elevation
differences rise towards Graham Land at the northern tip of
the Antarctic Peninsula, where topography is complex and data to constrain models at higher spatial resolutions. As a re-
highly variable at length scales similar to the satellite foot- sult, elevations derived from 5 km model fits will poorly sam-
print. ple the highly sloping terrain in this region when compared
In West Antarctica there is good agreement between the to the airborne laser. Additionally, in East Antarctica, eleva-
DEM and airborne measurements, particularly along the tion differences several tens of metres in magnitude over the
coastal margins of the Bellingshausen and Amundsen seas. Pensacola Mountains occur where high surface slopes and
In the Bryan and Eights coasts in the Bellingshausen Sea sec- nunataks complicate radar altimeter elevation retrievals.
tor, the median and rms difference are −1.72 and 10.40 m, At the Antarctic ice shelves, the DEM also compares
respectively. At Pine Island and Thwaites glaciers, and their favourably to the airborne elevation data, with median and
surrounding drainage area, the median difference is −1.02 m, rms differences of −0.42 and 14.31 m, respectively. Differ-
and the rms difference is 10.58 m. Further inland towards ences are most pronounced near to grounding lines where
Marie Byrd Land, the median and rms differences are 0.20 tidal effects are relatively large and where the terrain is gen-
and 5.27 m, respectively. erally more complex, and they are smallest in the interior of
In East Antarctica, the DEM compares well to the airborne the larger ice shelves, which are generally flat. At the Ross
data set inland in the plateau region where slopes are low, and Filchner-Ronne ice shelves, for example, the rms differ-
and the topography is well suited to satellite radar altimetry. ences are 3.93 and 3.54 m, respectively – considerably lower
Along the George V coast and in George V Land, the me- than the continental average.
dian difference is −0.68 m, and the rms difference is 6.52 m. Overall, the median and rms differences between the DEM
Over Totten Glacier and its catchment area, the median and and airborne measurements are −0.30 and 13.50 m, respec-
rms difference are −0.39 and 16.15 m, respectively. In this tively, and 99 % of the data agree to within 45 m. In addi-
region, there is good agreement with airborne elevations both tion to temporal mismatch, possible explanations for resid-
inland towards Dome C and over Totten Glacier itself. On the ual elevation differences include differences in the satellite
eastern flank of Law Dome, biases of several tens of metres and airborne altimeter footprint sizes and scattering horizons,
between the DEM and the airborne data coincide with grid as well as errors in the individual data sets themselves. Al-
cells derived from 5 km model fits, where there is insufficient though generally small, biases between the DEM and the air-
www.the-cryosphere.net/12/1551/2018/ The Cryosphere, 12, 1551–1562, 20181558 T. Slater et al.: A new digital elevation model of Antarctica derived from CryoSat-2 altimetry
Table 2. Statistics of the comparison between observed and interpolated DEM grid cells and airborne elevation measurements for individual
Antarctic regions and mode mask areas. In total, only 5 and 2 % of DEM elevation values are obtained through interpolation for the ice sheet
and ice shelves, respectively.
Observed Interpolated
Region Number of Median rms Number of Median rms
compared difference difference compared difference difference
grid cells (m) (m) grid cells (m) (m)
Ice sheet 230 165 −0.27 13.36 32 933 25.37 138.62
Ice shelves 40 081 −0.42 14.31 4772 1.20 30.96
Antarctic Peninsula 6820 −1.12 22.40 7473 82.21 191.07
West Antarctica 60 452 −0.86 11.43 8783 11.78 96.15
East Antarctica 162 893 −0.17 13.60 14 679 19.62 117.77
LRM 73 867 0.26 7.15 1683 6.51 41.70
SARIn (ice sheet only) 156 298 −0.82 15.45 31 250 28.65 141.97
Total 270 246 −0.30 13.50 37 655 19.84 131.13
(a) (b)
5 60
CryoSat-2 DEM
4 CryoSat-2 DEM corrected
50 Bamber et al., 2009
Bedmap2 (Fretwell et al., 2013)
Median difference (m)
3 Helm et al., 2014
RMS difference (m)
40
2
30
1
20
0
-1 10
-2 0
Ice Shelves Peninsula EAIS WAIS Ice Shelves Peninsula EAIS WAIS
Figure 8. (a) Median and (b) rms differences between airborne elevation measurements calculated over the ice shelves, Antarctic Peninsula,
West Antarctica (WAIS), and East Antarctica (EAIS) for the new CryoSat-2 DEM presented in this report, and three publicly available
Antarctic DEMs. CryoSat-2 DEM comparisons with the elevation change correction applied (Table 2) are plotted as grey bars.
borne data are notably high in several isolated regions, in- by CryoSat-2 in either LRM or SARIn modes (Table 3). In
cluding the upstream catchments of the Byrd Glacier flow- contrast, agreement between DEM and IceBridge elevations
ing from East Antarctica into the Ross Ice Shelf, the Re- in regions of lower surface slope (< 0.5◦ ) – which represent
covery Glacier flowing from East Antarctica into the Filch- the majority of the ice sheet – falls typically in the range
ner Ice Shelf, and the Foundation Ice Stream in the Pen- 5 to 10 m in either operating mode (Table 3). Combining
sacola Mountains (see Fig. 7). In each of these locations, the slope-dependent errors (Table 3) and the distribution of
surface slopes are high (exceeding 1◦ ) and CryoSat-2 oper- slopes within the LRM and SARIn mode masks, we estimate
ates in low-resolution mode (see Fig. 5). To illustrate this in the average uncertainty of the observed DEM to be 9.5 m.
more detail, we compare elevation recorded along two RLA
tracks falling within the LRM zone (Fig. 9): one at Byrd 3.2 Comparison of DEM to airborne elevation
Glacier where slopes are high and undulating, and another measurements: interpolated grid cells
600 km northward in Victoria Land where slopes are low and
smooth. Along these tracks, elevation differences of approx- A small proportion (5 %) of the DEM is estimated by or-
imately 20 m occur where the terrain undulates rapidly, be- dinary kriging, and we assess the accuracy of this method
cause CryoSat-2 undersamples the topographic depressions by comparing airborne elevation measurements residing in
when operating in LRM. Despite being well sampled by a DEM grid cell containing no data with the interpolated
the airborne laser altimeter data set, regions of high sur- value (Table 2). Predictably, our interpolated DEM values
face slopes represent a small fraction of the area surveyed deviate more from the airborne elevation measurements in ar-
eas of high slope and complex terrain, where internal tracker
The Cryosphere, 12, 1551–1562, 2018 www.the-cryosphere.net/12/1551/2018/T. Slater et al.: A new digital elevation model of Antarctica derived from CryoSat-2 altimetry 1559
Table 3. rms differences between observed DEM grid cell airborne
elevation measurements for four slope bands, separated into the
LRM and SARIn mode mask areas for the Antarctic ice sheet. The
area of each region represented by the four slope bands is also pro-
vided.
LRM SARIn
Slope (◦ ) rms LRM area rms SARIn area
difference coverage difference coverage
(m) (km2 ) (m) (km2 )
0–0.25 4.90 5 481 579 6.37 1 373 258
0.25–0.5 11.24 975 624 8.54 1 338 826
0.5–0.75 19.85 143 420 13.50 775 083
> 0.75 29.59 93 660 24.26 1 551 836
losses occur and data coverage is reduced. This is true in
particular for the ice sheet margins and the Antarctic Penin-
sula, where there is little spatial correlation over the 10, 25,
and 50 km search distances we have chosen for the interpo-
lation, and limited data coverage available for sampling. At
the Antarctic Peninsula, where the majority of interpolated
Figure 9. CryoSat-2 LRM and IceBridge RLA elevation profiles
grid cells are located in the bare-rock regions on the north
for 100 km flight path sections obtained in (a) Victoria Land, where
coasts of Graham and Palmer Land, the median and rms dif-
surface slopes are low, and (b) inland from Byrd Glacier, where
ference are 82.21 and 191.07 m, respectively. Similarly, in surface slopes are high. Elevation differences (CS2 DEM–airborne)
East Antarctica, where the median and rms difference are are plotted in blue to the right-hand scale. (Inset) Locations of
19.62 and 117.77 m, respectively, interpolated grid cells are RLA flight paths, with the profile section highlighted in red. The
primarily found in the rugged, bare-rock terrain across the LRM/SARIn mode mask boundary is shown as a dashed line.
Transantarctic Mountains, the Victory Mountains in Victoria
Land, and the mountain ranges in Oates Land.
The largest interpolation errors are located in grid cells at
the boundary of the ice sheet along the margins, as data gaps and ICESat data (Bamber et al., 2009), and a DEM gener-
are filled through extrapolation from data inland rather than ated from CryoSat-2 data (Helm et al., 2014) (for difference
interpolation between known values. Over higher-elevation maps see Fig. S1 of Supplement). To ensure an equivalent
regions with relatively smooth topography it is more reason- comparison data set, we only use airborne elevation measure-
able to assume spatial correlation over interpolation distances ments which reside in an observed grid cell of the presented
of 10 to 50 km, and our chosen interpolation method is more CryoSat-2 DEM (see Fig. 7). For all four DEMs we use the
reliable. Within the LRM zone, the median and rms differ- same evaluation method as described in Sect. 2.4. From the
ence are 6.51 and 41.70 m, respectively. We note that, be- calculated median and root mean square differences, the new
cause the elevation rate is unknown where there is no model CryoSat-2 DEM we present here is comparable to, or an im-
solution, we have not corrected for temporal changes in el- provement upon, currently available DEMs in all four regions
evation between the acquisition periods of the two data sets (Fig. 8). In areas of high rates of elevation change, it is worth
within our evaluation of interpolated grid cells. As a result, noting that all four DEMs will exhibit larger biases due to
the reported values represent an upper bound of the eleva- real changes in surface elevation between the acquisition pe-
tion difference which includes errors due to both interpola- riods of the respective data sets and that these differences
tion and elevation change – if present within an interpolated may be larger in the DEMs containing older ERS-1 and ICE-
grid cell. Sat data (Bedmap2; Bamber et al., 2009).
Similarly, median and root mean square differences cal-
3.3 Comparison of currently available DEMs culated with respect to surface slope for each DEM within
the Antarctic ice sheet (Fig. 10) further illustrate the im-
We compare the accuracy of the new CryoSat-2 DEM over provement offered by the new CryoSat-2 DEM and the slope-
the ice shelves, Antarctic Peninsula, West Antarctica, and dependency of DEM accuracy. We limit this analysis to re-
East Antarctica with three other publicly available Antarc- gions where surface slopes are lower than 1.5◦ , which ac-
tic DEMs: Bedmap2 (Fretwell et al., 2013), a DEM gen- counts for approximately 94 % of the Antarctic ice sheet area
erated from European Remote-Sensing Satellite–1 (ERS-1) north of 88◦ S. In addition, we note that the spatial distri-
www.the-cryosphere.net/12/1551/2018/ The Cryosphere, 12, 1551–1562, 20181560 T. Slater et al.: A new digital elevation model of Antarctica derived from CryoSat-2 altimetry
10
(a) altimeter elevation measurements, acquired over a contem-
poraneous time period and in a wide range of locations across
Median difference (m)
the Antarctic ice sheet and ice shelves. From a comparison at
5
grid cells acquired in both data sets, the median and rms dif-
ference between the DEM and airborne data are −0.30 and
0
13.50 m, respectively. The largest elevation differences oc-
cur in areas of high slope and where CryoSat-2 operates in
-5
0 0.25 0.5 0.75 1 1.25 1.5 low-resolution mode, where the altimeter ranges to the peaks
Slope (°) of undulating terrain and undersamples troughs. Using the
(b)
50 slope-dependent uncertainties and the wider distribution of
slopes, we estimate the overall accuracy of the DEM to be
RMS difference (m)
CryoSat-2 DEM
40 CryoSat-2 DEM corrected
Bamber et al., 2009
Bedmap2 (Fretwell et al., 2013)
Helm et al., 2014 9.5 m where elevations are formed from satellite data alone.
30
In areas where the DEM is interpolated, the median and rms
20 differences rise to 19.84 and 131.13 m, respectively. Through
10 comparisons to an equivalent validation data set in four indi-
0 vidual Antarctic regions, we find the new CryoSat-2 DEM
0 0.25 0.5 0.75 1 1.25 1.5 to be comparable to, or an improvement upon, three publicly
Slope (°)
available and widely used Antarctic DEMs.
Figure 10. (a) Median and (b) rms differences between airborne el-
evation measurements calculated for grid cells within the Antarctic
ice sheet, binned with respect to surface slope at a bin size of 0.05◦ Data availability. The new CryoSat-2 DEM will be made
for the new CryoSat-2 DEM presented in this report, and the three freely available to users via the Centre for Polar Observation
publicly available Antarctic DEMs used for comparison. CryoSat-2 and Modelling data portal (http://www.cpom.ucl.ac.uk/csopr/)
DEM differences with the elevation change correction applied are and via the European Space Agency (ESA) CryoSat Op-
plotted in grey. erational Portal (https://earth.esa.int/web/guest/missions/
esa-operational-eo-missions/cryosat).
CryoSat-2 data are available for download from ESA (https://
bution of the airborne data set used for comparison within earth.esa.int/web/guest/-/cryosat-products; ESA, 2018).
the grounded ice sheet preferentially samples regions of high IceBridge airborne altimetry data are available for download
slope and low elevation, and does not reflect the overall ele- from the National Snow and Ice Data Center (NSIDC) (https://
nsidc.org/icebridge/portal; Krabill, 2016; Blankenship et al., 2013).
vation and slope distributions of the Antarctic ice sheet. Ap-
proximately 60 % of DEM grid cells overflown by IceBridge
survey craft have an elevation of less than or equal to 1000 m,
and 43 % have a surface slope greater than 0.5◦ . In compar- The Supplement related to this article is available online
ison, approximately 15 and 22 % of the Antarctic ice sheet at https://doi.org/10.5194/tc-12-1551-2018-supplement.
area have elevations of less than 1000 m and slopes greater
than 0.5◦ , respectively.
Although another recent DEM of Antarctica (Fei et al.,
2017) formed using 1.7 × 107 elevation measurements ac- Competing interests. The authors declare that they have no conflict
quired by CryoSat-2 between 2012 and 2014 is not avail- of interest.
able for direct assessment, it has a reported accuracy of ap-
proximately 1 m for the high-elevation region at the Domes,
4 m for the ice shelves, and over 150 m for mountainous and Acknowledgements. This work was led by the NERC Centre
for Polar Observation and Modelling, supported by the Natural
coastal areas.
Environment Research Council (NERC) (cpom300001), with
the support of grant 4000107503/13/I-BG. We acknowledge the
4 Conclusions European Space Agency (ESA) for the provision of CryoSat-2 data,
ESA’s Antarcic_Ice Sheet_cci, and the National Snow and Ice Data
We present a new DEM of Antarctica derived from a spatio- Center (NSIDC) for the provision of IceBridge airborne altimetry
data. We also acknowledge the authors of the three digital elevation
temporal analysis of CryoSat-2 data acquired between
models used in this study, all of which are freely available online.
July 2010 and July 2016. The DEM is posted at a modal res- Thomas Slater is funded through the NERC Ice Sheet Stability
olution of 1 km and contains an elevation measurement in 94 (iSTAR) Programme and NERC grant number NE/J005681/1.
and 98 % of ice sheet and ice shelf grid cells, respectively; Anna E. Hogg is funded from the European Space Agency’s sup-
elevation in a further 5 % of the domain is estimated via or- port to Science Element programme and an independent research
dinary kriging. We evaluate the accuracy of the DEM through fellowship (4000112797/15/I-SBo). We thank the editor and three
comparison to an extensive independent set of airborne laser anonymous reviewers for their comments, which helped improve
The Cryosphere, 12, 1551–1562, 2018 www.the-cryosphere.net/12/1551/2018/T. Slater et al.: A new digital elevation model of Antarctica derived from CryoSat-2 altimetry 1561
the manuscript. CryoSat-2 altimetry data, Chinese J. Geophys.-CH, 60, 231–243,
https://doi.org/10.1002/cjg2.30041, 2017.
Edited by: Bert Wouters Flament, T. and Rémy, F.: Dynamic thinning of Antarctic glaciers
Reviewed by: three anonymous referees from along-track repeat radar altimetry, J. Glaciol., 58, 830–840,
https://doi.org/10.3189/2012JoG11J118, 2012.
Fretwell, P., Pritchard, H. D., Vaughan, D. G., Bamber, J. L., Bar-
rand, N. E., Bell, R., Bianchi, C., Bingham, R. G., Blanken-
References ship, D. D., Casassa, G., Catania, G., Callens, D., Conway, H.,
Cook, A. J., Corr, H. F. J., Damaske, D., Damm, V., Ferracci-
Armitage, T., Wingham, D., and Ridout, A.: Meteoro- oli, F., Forsberg, R., Fujita, S., Gim, Y., Gogineni, P., Griggs,
logical origin of the static crossover pattern present J. A., Hindmarsh, R. C. A., Holmlund, P., Holt, J. W., Jacobel,
in low-resolution-mode CryoSat-2 data over Central R. W., Jenkins, A., Jokat, W., Jordan, T., King, E. C., Kohler,
Antarctica, IEEE Geosci. Remote S., 11, 1295–1299, J., Krabill, W., Riger-Kusk, M., Langley, K. A., Leitchenkov,
https://doi.org/10.1109/LGRS.2013.2292821, 2014. G., Leuschen, C., Luyendyk, B. P., Matsuoka, K., Mouginot,
Bamber, J. L., Gomez-Dans, J. L., and Griggs, J. A.: A new 1 km J., Nitsche, F. O., Nogi, Y., Nost, O. A., Popov, S. V., Rignot,
digital elevation model of the Antarctic derived from combined E., Rippin, D. M., Rivera, A., Roberts, J., Ross, N., Siegert,
satellite radar and laser data – Part 1: Data and methods, The M. J., Smith, A. M., Steinhage, D., Studinger, M., Sun, B.,
Cryosphere, 3, 101–111, https://doi.org/10.5194/tc-3-101-2009, Tinto, B. K., Welch, B. C., Wilson, D., Young, D. A., Xiangbin,
2009. C., and Zirizzotti, A.: Bedmap2: improved ice bed, surface and
Blankenship, D. D.,Young, D. D., Kempf, S., Roberts, J. L., van thickness datasets for Antarctica, The Cryosphere, 7, 375–393,
Ommen, T., Forsberg, R., Siegert, M. J., Palmer, S. J., and https://doi.org/10.5194/tc-7-375-2013, 2013.
Dowdeswell, J. A.: IceBridge Riegl Laser Altimeter L2 Ge- Griggs, J. A. and Bamber, J. L.: A new 1 km digital elevation model
olocated Surface Elevation Triplets, NASA DAAC at the Na- of Antarctica derived from combined radar and laser data – Part
tional Snow and Ice Data Center, Boulder, Colorado USA, 2: Validation and error estimates, The Cryosphere, 3, 113–123,
https://doi.org/10.5067/JV9DENETK13E, 2013 (data available https://doi.org/10.5194/tc-3-113-2009, 2009.
at: https://nsidc.org/icebridge/portal/, last access: January 2017). Helm, V., Humbert, A., and Miller, H.: Elevation and elevation
Cook, A. J., Murray, T., Luckman, A., Vaughan, D. G., and Bar- change of Greenland and Antarctica derived from CryoSat-
rand, N. E.: A new 100-m Digital Elevation Model of the 2, The Cryosphere, 8, 1539–1559, https://doi.org/10.5194/tc-8-
Antarctic Peninsula derived from ASTER Global DEM: meth- 1539-2014, 2014.
ods and accuracy assessment, Earth Syst. Sci. Data, 4, 129–142, Isaaks, E. H. and Srivastava, R. M.: An Introduction to Applied
https://doi.org/10.5194/essd-4-129-2012, 2012. Geostatistics, Oxford University Press, New York, 1989.
Cornford, S. L., Martin, D. F., Payne, A. J., Ng, E. G., Le Brocq, A. Khvorostovsky, K. S.: Merging and analysis of eleva-
M., Gladstone, R. M., Edwards, T. L., Shannon, S. R., Agosta, tion time series over Greenland ice sheet from satel-
C., van den Broeke, M. R., Hellmer, H. H., Krinner, G., Ligten- lite radar altimetry, IEEE T. Geosci. Remote, 50, 23–36,
berg, S. R. M., Timmermann, R., and Vaughan, D. G.: Century- https://doi.org/10.1109/TGRS.2011.2160071, 2012.
scale simulations of the response of the West Antarctic Ice Kitanitis, P. K.: Introduction to Geostatistics: Applications in Hy-
Sheet to a warming climate, The Cryosphere, 9, 1579–1600, drogeology, Cambridge University Press, Cambridge, 1997.
https://doi.org/10.5194/tc-9-1579-2015, 2015. Koenig, L., Martin, S., Studinger, M., and Sonntag, J.: Polar Air-
Davis, C. H.: A robust threshold retracking algorithm for borne Observations Fill Gap in Satellite Data, EOS T. Am. Geo-
measuring ice-sheet surface elevation change from satellite phys. Un., 91, 333–334, https://doi.org/10.1029/2010EO380002,
radar altimeters, IEEE T. Geosci. Remote, 35, 974–979, 2010.
https://doi.org/10.1109/36.602540, 1997. Korona, J., Berthier, E., Bernard, M., Rémy, F., and Thouvenot, E.:
Davis, C. H. and Ferguson, A. C.: Elevation change of SPIRIT. SPOT 5 stereoscopic survey of Polar Ice: Reference
the Antarctic ice sheet, 1995–2000, from ERS-2 satellite Images and Topographies during the fourth International Po-
radar altimetry, IEEE T. Geosci. Remote, 42, 2437–2445, lar Year (2007–2009), ISPRS J. Photogramm., 64, 204–212,
https://doi.org/10.1109/TGRS.2004.836789, 2004. https://doi.org/10.1016/j.isprsjprs.2008.10.005, 2009.
DiMarzio, J., Brenner, A., Schutz, R., Shuman, C. A., and Krabill, W. B.: IceBridge ATM L2 Icessn Elevation, Slope, and
Zwally, H. J.: GLAS/ICESat 500 m laser altimetry digital ele- Roughness Version 2, NASA DAAC at the National Snow
vation model of Antarctica, National Snow and Ice Data Center, and Ice Data Center, Digital Media, Boulder, Colorado, USA,
Digital Media, Boulder, Colorado, USA, 2007. https://doi.org/10.5067/CPRXXK3F39RV, 2016 (data available
ESA: CryoSat-2 Product Handbook, ESRIN-ESA and Mullard at: https://nsidc.org/icebridge/portal/, last access: January 2017).
Space Science Laboratory, University College London, Krabill, W. B., Hanna, E., Huybrechts, P., Abdalati, W., Cappe-
available at: http://emits.sso.esa.int/emits-doc/ESRIN/7158/ len, J., Csatho, B., Frederick, E., Manizade, S., Martin, C.,
CryoSat-PHB-17apr2012.pdf (last access: 2 October 2017), Sonntag, J., Swift, R., Thomas, R., and Yungel, J.: Greenland
2012. Ice Sheet: Increased Coastal Thinning, Geophys. Res. Lett., 31,
ESA: CryoSat-2 Ice Level 2 data, available at: https://earth.esa.int/ L24402, https://doi.org/10.1029/2004GL021533, 2004.
web/guest/-/cryosat-products (last access: October 2016), 2018. Kwok, R., Cunningham, G. F, Manizade, S. S., and Krabill, W. B.:
Fei, L., Feng, X., Sheng-Kai, Z., Dong-Chen, E., Xiao, C., Arctic sea ice freeboard from IceBridge acquisitions in 2009:
Wei-Feng, H., Le-Xian, Y., and Yao-Wen, Z.: DEM Devel-
opment and precision analysis for Antarctic ice sheet using
www.the-cryosphere.net/12/1551/2018/ The Cryosphere, 12, 1551–1562, 20181562 T. Slater et al.: A new digital elevation model of Antarctica derived from CryoSat-2 altimetry Estimates and comparisons with ICESat, J. Geophys. Res., 117, Sutterly, T. C., Velicogna, I., Rignot, E., Mouginot, J., Flament, T., C02018, https://doi.org/10.1029/2011JC007654, 2012. van den Broeke, M. R., van Wessem, J. M., and Reijmer, C. H.: Levinsen, J. F., Howat, I. M., and Tscherning, C. C.: Improving Mass loss of the Amundsen Sea Embayment of West Antarc- maps of ice-sheet surface elevation change using combined laser tica from four independent techniques, Geophys. Res. Lett., 41, altimeter and stereoscopic elevation model data, J. Glaciol., 59, 8421–8428, https://doi.org/10.1002/2014GL061940, 2014. 524–532, https://doi.org/10.3189/2013JoG12J114, 2013. Wingham, D. J. and Wallis, D. W.: The Rough Surface Im- Liu, H., Jezek, K. C., Li, B., and Zhao, Z.: Radarsat Antarctic Map- pulse Response of a Pulse-Limited Altimeter With an Ellip- ping Project Digital Elevation Model Version 2, National Snow tical Antenna Pattern, IEEE Antenn. Wirel. Pr., 9, 232–235, and Ice Data Center, Digital Media, Boulder, Colorado, USA, https://doi.org/10.1109/LAWP.2010.2046471, 2010. 2001. Wingham, D. J., Rapley, C. G., and Griffiths, H.: New techniques McMillan, M., Shepherd, A., Sundal, A., Briggs, K., Muir, A., in satellite altimeter tracking systems, in: Proceedings of the Ridout, A., Hogg, A., and Wingham, D.: Increased ice losses IGARSS Symposium, vol. SP-254, edited by: Guyenne, T. D. from Antarctica detected by CryoSat-2, Geophys. Res. Lett., 41, and Hunt, J. J., European Space Agency, Zurich, Septem- 3899–3905, https://doi.org/10.1002/2014GL060111, 2014. ber 1986, 1339–1344, 1986. Mouginot, J., Scheuchl, B., and Rignot, E.: Mapping of Ice Mo- Wingham, D. J., Ridout, A. J., Scharroo, R., Arth- tion in Antarctica Using Synthetic-Aperture Radar Data, Remote ern, R. J., and Shum, C. K.: Antarctic Elevation Sens.-Basel, 4, 2753–2767, https://doi.org/10.3390/rs4092753, Change from 1992 to 1996, Science, 282, 456–458, 2012. https://doi.org/10.1126/science.282.5388.456, 1998. Rignot, E., Mouginot, J., and Scheuchl, B.: Ice Flow Wingham, D. J., Phalippou, L., Mavrocordatos, C., and Wal- of the Antarctic Ice Sheet, Science, 333, 1427–1430, lis, D.: The Mean Echo Cross Product From a Beamform- https://doi.org/10.1126/science.1208336, 2011a. ing Interferometric Altimeter and Their Application to Eleva- Rignot, E., Velicogna, I., van den Broeke, M. R., Monaghan, A., tion Measurement, IEEE T. Geosci. Remote, 42, 2305–2323, and Lenaerts, J. T. M.: Acceleration of the contribution of the https://doi.org/10.1109/TGRS.2004.834352, 2004. Greenland and Antarctic ice sheets to sea level rise, Geophys. Wingham, D., Francis, C. R., Baker, S., Bouzinac, C., Cullen, R., Res. Lett., 38, L05503, https://doi.org/10.1029/2011GL046583, de Chateau-Thierry, P., Laxon, S. W., Mallow, U., Mavro- 2011b. cordatos, C., Phalippou, L., Ratier, G., Rey, L., Rostan, F., Ritz, C., Edwards, T. L., Durand, G., Payne, A. J., Peyaud, V., and Viau, P., and Wallis, D.: CryoSat: a mission to determine the fluc- Hindmarsh, R. C. A.: Potential sea-level rise from Antarctic ice- tuations in Earth’s land and marine ice fields, Adv. Space Res., sheet instability constrained by observations, Nature, 528, 115– 37, 841–871, https://doi.org/10.1016/j.asr.2005.07.027, 2006. 118, https://doi.org/10.1038/nature16147, 2015. Young, D. A., Kempf, S. D., Blankenship, D. D., Holt, J. W., and Shepherd, A., Ivins, E. R., Geruo, A., Barletta, V. R., Bent- Morse, D. L.: New airborne laser altimetry over the Thwaites ley, M. J., Bettadpur, S., Briggs, K. H., Bromwich, D. H., catchment, West Antarctica, Geochem. Geophy. Geosy., 9, Forsberg, R., Galin, N., Horwath, M., Jacobs, S., Joughin, I., Q06006, https://doi.org/10.1029/2007GC001935, 2008. King, M. A., Lenaerts, J. T. M., Li, J., Ligtenberg, S. R. M., Luck- Zwally, H. J., Giovinetto, M. B., Beckley, M. A., and man, A., Luthcke, S. B., McMillan, M., Meister, R., Milne, G., Saba, J. J. L.: Antarctic and Greenland Drainage Sys- Mouginot, J., Muir, A., Nicolas, J. P., Paden, J., Payne, A. J., tems, available at: http://icesat4.gsfc.nasa.gov/cryo_data/ant_ Pritchard, H., Rignot, E., Rott, H., Sorensen, L. S., Scam- grn_drainage_systems.php (last access: 15 May 2017), 2012. bos, T. A., Scheuchl, B., Schrama, E. J. O., Smith, B., Sun- dal, A. V., van Angelen, J. H., van de Berg, W. J., van den Broeke, M. R., Vaughan, D. G., Velicogna, I., Wahr, J., White- house, P. L., Wingham, D. J., Yi, D., Young, D., and Zwally, H. J.: A reconciled estimate of ice-sheet mass balance, Science, 338, 1183–1189, https://doi.org/10.1126/science.1228102, 2012. The Cryosphere, 12, 1551–1562, 2018 www.the-cryosphere.net/12/1551/2018/
You can also read
